Keyword determinations from voice data

Abstract

Topics of potential interest to a user, useful for purposes such as targeted advertising and product recommendations, can be extracted from voice content produced by a user. A computing device can capture voice content, such as when a user speaks into or near the device. One or more sniffer algorithms or processes can attempt to identify trigger words in the voice content, which can indicate a level of interest of the user. For each identified potential trigger word, the device can capture adjacent audio that can be analyzed, on the device or remotely, to attempt to determine one or more keywords associated with that trigger word. The identified keywords can be stored and/or transmitted to an appropriate location accessible to entities such as advertisers or content providers who can use the keywords to attempt to select or customize content that is likely relevant to the user.

What is claimed is: 
     
       1. A computer-implemented method, comprising:
 under the control of a computing device configured with executable instructions,
 receiving voice content corresponding to a user; 
 determining that a first trigger word is included in the voice content; 
 determining a second trigger word based at least upon two or more of a past behavior data of the user, a received update from another source, or another trigger word associated with the computing device; 
 determining a keyword that is at least one of an object of the second trigger word or an object of the first trigger word; and 
 determining content to be provided to the user based at least in part upon the keyword, the keyword being indicative of a topic of potential interest to the user. 
 
 
     
     
       2. The computer-implemented method of  claim 1 , wherein the first trigger word corresponds to at least one of two or more trigger words stored by the computing device or at least one remote data store. 
     
     
       3. The computer-implemented method of  claim 2 , wherein determining that the first trigger word is included in the voice content includes comparing the first trigger word with the two or more trigger words and variants of the two or more trigger words. 
     
     
       4. The computer-implemented method of  claim 1 , wherein determining the keyword includes at least one of analyzing at least one portion of the voice content on the computing device or transmitting the at least one portion of the voice content to another computing device to analyze the portion of the voice content. 
     
     
       5. The computer-implemented method of  claim 1 , wherein determining the keyword includes using at least one speech recognition process to identify words in the voice content. 
     
     
       6. The computer-implemented method of  claim 5 , wherein determining the keyword includes analyzing at least one portion of the voice content using at least one text analytics algorithm. 
     
     
       7. The computer-implemented method of  claim 1 , further comprising:
 storing the voice content temporarily in one or more registers of the computing device, the one or more registers configured to store data received from the computing device. 
 
     
     
       8. The computer-implemented method of  claim 1 , further comprising:
 identifying a source of at least one portion of the voice content that corresponds to the first trigger word, and 
 causing the at least one portion of the voice content to be analyzed based at least upon an identity of the source. 
 
     
     
       9. The computer-implemented method of  claim 8 , wherein identifying the source of the at least one portion of the voice content comprises using at least one of voice recognition, facial recognition, motion detection, or geographic location detection. 
     
     
       10. The computer-implemented method of  claim 1 , further comprising:
 causing the keyword to be associated with the user. 
 
     
     
       11. The computer-implemented method of  claim 10 , further comprising:
 causing additional information to be stored with the keyword, the additional information including at least one of a time stamp, geographic coordinates, identity information of the user or a source of the keyword, context of the voice content, information regarding to level or type of the topic of interest, or a priority of the keyword. 
 
     
     
       12. The computer-implemented method of  claim 11 , further comprising:
 assigning the priority of the keyword or a weight to the keyword based at least upon one of a number of repetitions in the voice content or closeness in time according to the time stamp. 
 
     
     
       13. The computer-implemented method of  claim 1 , wherein the determined content includes one of advertising, product information, or at least one recommendation. 
     
     
       14. A system, comprising:
 at least one processor; and 
 memory including instructions that, when executed by the at least one processor, cause the system to:
 receive voice content corresponding to a user; 
 determine that a first trigger word is included in the voice content; 
 determine a second trigger word based at least upon two or more of a past behavior data of the user, a received update from another source, or another trigger word associated with the system; 
 determine a keyword that is at least one of an object of the second trigger word or an object of the first trigger word; and 
 determine content to be provided to the user based at least in part upon the keyword, the keyword being indicative of a topic of potential interest to the user. 
 
 
     
     
       15. The system of  claim 14 , wherein the instructions when executed further cause the system to:
 one audio generation component operable to transmit audio generated from data from the system; wherein the determined content is conveyed to the user via the one audio generation component or a display of the system. 
 
     
     
       16. The system of  claim 15 , wherein the instructions when executed further cause the system to:
 determine that the user is engaged in a telecommunication call using the system; and 
 convey the determined content to the user at a controlled volume via the one audio generation component. 
 
     
     
       17. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computer system, cause the computer system to:
 receive voice content corresponding to a user; 
 determine that a first trigger word is included in the voice content; 
 determine a second trigger word based at least upon two or more of a past behavior data of the user, a received update from another source, or another trigger word associated with the computing system; 
 determine a keyword that is at least one of an object of the second trigger word or an object of the first trigger word; and 
 determine content to be provided to the user based at least in part upon the keyword, the keyword being indicative of a topic of potential interest to the user. 
 
     
     
       18. The non-transitory computer-readable storage medium of  claim 17 , wherein the instructions when executed further cause the system to:
 identify a source of at least one portion of the voice content that corresponds to the first trigger word, and 
 cause the at least one portion of the voice content to be analyzed based at least upon an identity of the source. 
 
     
     
       19. The non-transitory computer-readable storage medium of  claim 17 , wherein the instructions when executed further cause the system to:
 cause additional information to be stored with the keyword, the additional information including at least one of a time stamp, geographic coordinates, identity information of the user or a source of the keyword, context of the voice content, information regarding to level or type of the topic of interest, or a priority of the keyword. 
 
     
     
       20. The system of  claim 14 , wherein the instructions when executed further cause the system to:
 identify a source of at least one portion of the voice content that corresponds to the first trigger word, and 
 cause the at least one portion of the voice content to be analyzed based at least upon an identity of the source.